2 min

Okta has announced a new, risk-based authentication method that uses machine learning for automated detection of and response to identity-based attacks.

According to the company, risk-based authentication provides a “panoramic view” of an organisation’s employees, partners and customers. It should also be a means of transparent security controls. It allows companies to automate their security and implement stronger authentication and recovery techniques.

According to Okta’s CEO and co-founder Todd McKinnon, the company’s new solution will make security “a strong point for the user experience, while maintaining protection where it’s needed”.

“The combination of the individualised context of Machine Learning with insights based on the vast amount of data from Oktas’ extensive ecosystem of customers, integrations and authentications, creates a high level of intelligence that enables transparent and at the same time workable security.

Individual profiles

Each user is unique in which applications and devices he uses and the locations from which he wants to access them. Okta therefore used machine learning to create an authentication context at the user level. Individual profiles are created, which are further updated over time thanks to machine learning. In this way, a thorough understanding of how users deal with and access the technology they need is created.

When combined with information from Okta’s ThreatInsight – risk signals perceived by the security company in its global dataset, such as high-risk IP addresses – it is possible to make quantifiable and actionable risk assessments for each authentication attempt.

The solution also enables recovery measures for user-specific accounts, according to the company. In addition, a user verifies a possible security incident, which triggers the recovery workflow. And for users, the login experience remains simple by using Okta Adaptive Single Sign-On and Okta Adaptive Multi-Factor Authentication.


Okta has made the product available in beta. It is expected to be widely available soon as part of Adaptive Single Sign-On and Multi-Factor Authentication.

This news article was automatically translated from Dutch to give Techzine.eu a head start. All news articles after September 1, 2019 are written in native English and NOT translated. All our background stories are written in native English as well. For more information read our launch article.